The Empty Scorebook: Cricket Coverage's Biggest Error Sits in the Ledger Nobody Audits

The Empty Scorebook: Cricket Coverage's Biggest Error Sits in the Ledger Nobody Audits

April morning. The press box at a county ground, before ten, colder inside than out. Dew on the grass, the smell of wet earth in the air, and from the nets that stubborn, repeated sound — thud, thud, thud. I arrive two hours early; that habit has not changed in forty years. Clicker in the pocket, notebook in hand, and inside the front cover the numbered evidence index that travels with me on every trip.

That morning the scoring feed was down. The scoreboard sat at 0/0 for six overs and nobody in the press box filed a complaint. There was nobody there to file one. Four seats, two people, both freelance, both on a three-hour copy deadline.

I counted the deliveries, and the count told its own story. My clicker logged 214 balls that day. The newspaper record next morning carried 198. Twenty-six deliveries fell off a server and disappeared, and to this day nobody knows which twenty-six.

That is the biggest error in this trade. Not in a dropped catch, not in a DRS review — in the register nobody audits. The ground was empty; the microphones were not. And the microphones that stay live are usually reporting something nobody is verifying.

Context: who manufactures a cricket fact

Every cricket story sits on a supply chain, and the chain has three floors. At the top, international cricket: the ICC media wing, board press releases, the broadcaster's graphics team, two or three wire correspondents. On this floor every figure is entered in at least three places, so an error does not survive. The middle floor is franchise cricket: here the number is commercial, the club sells it, and so it is kept clean — even if clean is not the same as true.

The bottom floor is domestic cricket. Ranji knockouts, county cricket in April, regional tournaments, the women's domestic calendar, the under-19 and under-23 games whose scores appear on a board website two days later. Nobody sends a correspondent to this floor because nobody pays for one. Where there is no payment there is no audit. Where there is no audit, an error lives longest.

Two things have happened side by side over the past decade — broadcast rights values climbed on one side, and the coverage nobody buys shrank on the other. That is not a paradox, it is cause and effect. The year a board cashed a large cheque from a network was the same financial year it ended the contracts of three reporters covering its own domestic tournament. Cricket is now visible to the exact edge of the camera. And the camera does not go where the market does not.

Inside that structure a third thing has appeared, which I call the empty file. A report reaches me — of a match, a tour, a selection meeting. Every box looks filled, from headline to conclusion, but the list of facts is blank. Zero information points. When that happens, the first question an old reporter asks is not what happened in the match. It is where the match report got stuck.

Core: counting, documents, and the ten-match threshold

My method is no secret. It is just slow. I count deliveries — by spell, by bowler, by date. In the summer of 2026 I sat behind the nets at Lord's with a clicker and logged 1,186 deliveries in practice sessions and 1,412 in matches. That same June at Lord's, Rahul Dravid made 95 and Sourav Ganguly 131, both on debut, both in the same match. England won that series 1-0, and it is the series no highlights reel remembers.

The clicker stopped being equipment that day. It became a habit, and a habit keeps running when memory begins to lie. Four years later, in April 2026, when Hansie Cronje's confession broke and the Tehelka tapes followed in May, the desk wanted names inside forty-eight hours. I refused to supply names and turned back through fourteen months of notebooks — sixty-one matches, every field setting, every over that did not add up. Across twenty-two pages of chronology I printed only what carried two documents behind it. The rest went in the bin.

That decision cost me two scoops and bought me fifteen years of access. Since then a numbered evidence index has been stapled inside the front cover of every notebook I carry. My rule: before any allegation, a dated source log, and a second document that corroborates it. One document gives you a story. The second document tells you whose story it was.

What happened at Lord's in August 2026 was not a lesson that a player can be corrupt — that lesson arrived in 2026. The lesson was that the irregularity usually hides in the figure everyone treats as an ordinary day: a no-ball, a run-up's timing, an over rate that suddenly improves. A pattern becomes visible to a reporter only when the reporter is counting. I keep the beat before I keep the opinion, because an opinion can be sold into the market and a beat cannot.

There is a technical point cricket coverage forgets again and again, and it is format. A Test average, an ODI average and a T20 strike rate are not comparable. An economy rate that looks excellent in ODI cricket may be useless in the death overs of a T20. In a report with no format tag, every comparison is already invalid. I have read reports where a Test opening spell and the last over of an IPL match sat in the same paragraph and the reader was never told what had been sold to him.

Second, what I call the ten-match threshold. One innings cannot finish a player or create one. Ten matches is the minimum sample at which weather, pitch, opposition bowling and the toss stop shouting. Below that I still write — but observation only, never verdict. The modern deadline erases that distinction, and what survives the erasure is not cricket, it is reaction.

And now the part that troubles me most

When the empty file arrived — filled from headline to conclusion, and zero in the list of facts — my first reaction was that perhaps no match had taken place. That conclusion was wrong, and wrong because I had mistaken absence for event. A match that never happened and a match whose record was never made are different things, and the difference is the foundation of this trade. An empty file most likely means the extraction stopped somewhere in the middle: the source text was never ingested, or ingested and never recognised. That is a process failure, not a cricket event.

This is where the industry makes its largest mistake. A filled template reads as safe. Eighteen boxes out of eighteen answered, and the reader assumes the checking is done. But if not one of those eighteen holds a hard fact, the report is not a risk assessment, it is the illusion of one. That is the most dangerous document I know — the one where 'no risk found' and 'no risk assessable' look identical. The man on the desk cannot tell them apart, and by next morning the same sentence has travelled onward.

A press-box habit operates here too, and I apply it to myself. At sixty, after this long on the beat, I am on first-name terms with the boards I cover. That access makes the work easier and the thinking lazier. So I write beside every fact: who supplied it, and what they gain from it appearing. If the source has a stake, the story needs a second source with the opposite stake. That is not suspicion, it is bookkeeping.

Contrarian: more data does not mean more truth

The industry's most comfortable belief is that more information produces more truth. Tournament cycles deepen the belief, because every channel has data to fill graphics and every desk has space to fill with copy. But a pile of data is not an audit. This season I saw at least four figures printed three different ways in three places, and nobody asked which was right. Two of the three carried the number that had first appeared in a press release. One source, counted three times, treated as three independent confirmations.

The truth is that verification is expensive and cricket's economy has no price for it. No board pays for verification. No network buys the correspondent who turns up two hours early. So verification is the first thing cut and the last thing remembered — usually when an error is large enough to require an apology.

I have only one remedy, and it is habitual rather than professional: publish the observation and the count immediately, hold the causal claim. Sequence the caution, do not apply it to everything. The moment I reverse that order I become a witness in my own court — and my credibility rests on never having printed the first claim before the second document arrived.

One thing must be added. When the information points in a report are blank, the correct conclusion is not 'nothing was found'. It is 'the check is incomplete', and that is itself a story. If that story goes unrecorded, the same empty file will look full next time, and nobody will even suspect it.

Takeaway: what to watch next

I will enter three things in my ledger. First, the population rate — how many information points a report carries and how many named entities are resolved; if that number is zero, the story is about process, not cricket. Second, a format tag on every figure — Test, ODI, T20 — because a number without a tag is not available for comparison. Third, a note beside every fact about what the source gains from it appearing.

If the register comes back empty, the news is not that nothing happened. The news is that nobody recorded it. And if even one correspondent this season starts asking where the figure on his desk came from, then next season those twenty-six deliveries may be found logged somewhere.

I went back to the nets to check what I remembered. The clicker is still in my pocket. The question is whether, the next time an empty file lands on my desk, I will print it empty — or write on the back of the page why it was never filled.